An essential aspect of human behavior is the ability to perform skilled and purposive movements. This ability relies on an intact neocortex, and in particular on different classes of projection neurons that convey neural information from the neocortex to downstream circuits involved in behavior. Two classes of cortical projection neurons are of particular interest - corticospinal neurons, projecting to the spinal cord, and corticostriatal neurons, projecting to the striatum bilaterally. While much has been learned from brain slice experiments about the cellular and synaptic properties of corticospinal and corticostriatal neurons, knowledge about their activity patterns in vivo in relation to volitional movements has been impeded due to a lack of easy to use methods for selectively recording from and manipulating the activity of these projection neurons. Here, building on prior ex vivo experiments, we propose a research program to develop new approaches to address this gap. In one aim, we will develop multiple complementary approaches to record the activities of corticospinal and corticostriatal neurons in the awake, moving mouse. In one set of studies we will use in vivo whole-cell recordings to record from motor cortex neurons at high sensitivity and temporal resolution, with post-hoc identification of cell types. In parallel studies we will infect corticospinal and corticostriatal neurons with recombinant rabies viruses carrying genetically encoded calcium indicators (GECIs), and detect activity either at high resolution using two- photon calcium imaging or at a population level using optical fibers. In our second aim, we will develop methods to manipulate the activities of corticospinal and corticostriatal neurons in the awake, moving mice. By either up- or down-regulate projection neuron activity using optogenetic or pharmacogenetic tools, we will be able to begin to address the question of how corticospinal and corticostriatal neurons' activities are causally related to movement parameters. The proposed research program is innovative and significant, we believe, because it will generate multiple new experimental paradigms for probing the in vivo functions of cortical projection neurons, not only in motor areas but across all of neocortex, and because these tools will be readily applicable to mouse models of neurological disease.